Learned features of antibody-antigen binding affinity.

Defining predictors of antigen-binding affinity of antibodies is valuable for engineering therapeutic antibodies with high binding affinity to their targets. However, this task is challenging owing to the huge diversity in the conformations of the complementarity determining regions of antibodies and the mode of engagement between antibody and antigen. In this study, we used the structural antibody database (SAbDab) to identify features that can discriminate high- and low-binding affinity across a 5-log scale. First, we abstracted features based on previously learned representations of protein-protein interactions to derive 'complex' feature sets, which include energetic, statistical, network-based, and machine-learned features. Second, we contrasted these complex feature sets with additional 'simple' feature sets based on counts of contacts between antibody and antigen. By investigating the predictive potential of 700 features contained in the eight complex and simple feature sets, we observed that simple feature sets perform comparably to complex feature sets in classification of binding affinity. Moreover, combining features from all eight feature-sets provided the best classification performance (median cross-validation AUROC and F1-score of 0.72). Of note, classification performance is substantially improved when several sources of data leakage (e.g., homologous antibodies) are not removed from the dataset, emphasizing a potential pitfall in this task. We additionally observe a classification performance plateau across diverse featurization approaches, highlighting the need for additional affinity-labeled antibody-antigen structural data. The findings from our present study set the stage for future studies aimed at multiple-log enhancement of antibody affinity through feature-guided engineering.

Complexity of Viral Epitope Surfaces as Evasive Targets for Vaccines and Therapeutic Antibodies.

The dynamic interplay between virus and host plays out across many interacting surfaces as virus and host evolve continually in response to one another. In particular, epitope-paratope interactions (EPIs) between viral antigen and host antibodies drive much of this evolutionary race. In this review, we describe a series of recent studies examining aspects of epitope complexity that go beyond two interacting protein surfaces as EPIs are typically understood. To structure our discussion, we present a framework for understanding epitope complexity as a spectrum along a series of axes, focusing primarily on 1) epitope biochemical complexity (e.g., epitopes involving N-glycans) and 2) antigen conformational/dynamic complexity (e.g., epitopes with differential properties depending on antigen state or fold-axis). We highlight additional epitope complexity factors including epitope tertiary/quaternary structure, which contribute to epistatic relationships between epitope residues within- or adjacent-to a given epitope, as well as epitope overlap resulting from polyclonal antibody responses, which is relevant when assessing antigenic pressure against a given epitope. Finally, we discuss how these different forms of epitope complexity can limit EPI analyses and therapeutic antibody development, as well as recent efforts to overcome these limitations.

A structural dynamic explanation for observed escape of SARS-CoV-2 BA.2 variant mutation S371L/F.

The SARS-CoV-2 Omicron sub-variants BA.1 and BA.2 have become the dominant variants worldwide due to enhanced transmissibility and immune evasion. In response to the rise of BA.1 and BA.2, two recent studies by Liu et al. and Iketani et al. provide a detailed analysis of loss of therapeutic antibody potency through evaluation of escape by pseudotyped viruses harboring BA.1 and BA.2 receptor binding domain (RBD) point mutations. Surprisingly, Liu et al. and Iketani et al. observed a profoundly broad escape effect for the individual mutations S371L and S371F. This result cannot be explained by known escape mechanisms of the SARS-CoV-2 RBD, and conflicts with existing computational and experimental escape measurements for S371 mutations performed on monomeric RBD. Through an examination of these conflicting datasets and a structural analysis of the antibodies assayed by Liu et al. and Iketani et al., we propose a mechanism to explain S371L/F escape according to a perturbation of spike trimer conformational dynamics that has not yet been described for any SARS-CoV-2 escape mutation. The proposed mechanism is relevant to Omicron and future variant surveillance as well as therapeutic antibody design.

Insights on the mutational landscape of the SARS-CoV-2 Omicron variant receptor-binding domain.

The Omicron variant features enhanced transmissibility and antibody escape. Here, we describe the Omicron receptor-binding domain (RBD) mutational landscape using amino acid interaction (AAI) networks, which are well suited for interrogating constellations of mutations that function in an epistatic manner. Using AAI, we map Omicron mutations directly and indirectly driving increased escape breadth and depth in class 1-4 antibody epitopes. Further, we present epitope networks for authorized therapeutic antibodies and assess perturbations to each antibody's epitope. Since our initial modeling following the identification of Omicron, these predictions have been realized by experimental findings of Omicron neutralization escape from therapeutic antibodies ADG20, AZD8895, and AZD1061. Importantly, the AAI predicted escape resulting from indirect epitope perturbations was not captured by previous sequence or point mutation analyses. Finally, for several Omicron RBD mutations, we find evidence for a plausible role in enhanced transmissibility via disruption of RBD-down conformational stability at the RBDdown-RBDdown interface.

Conserved topology of virus glycoepitopes presents novel targets for repurposing HIV antibody 2G12.

Complex glycans decorate viral surface proteins and play a critical role in virus-host interactions. Viral surface glycans shield vulnerable protein epitopes from host immunity yet can also present distinct "glycoepitopes" that can be targeted by host antibodies such as the potent anti-HIV antibody 2G12 that binds high-mannose glycans on gp120. Two recent publications demonstrate 2G12 binding to high mannose glycans on SARS-CoV-2 and select Influenza A (Flu) H3N2 viruses. Previously, our lab observed 2G12 binding and functional inhibition of a range of Flu viruses that include H3N2 and H1N1 lineages. In this manuscript, we present these data alongside structural analyses to offer an expanded picture of 2G12-Flu interactions. Further, based on the remarkable breadth of 2G12 N-glycan recognition and the structural factors promoting glycoprotein oligomannosylation, we hypothesize that 2G12 glycoepitopes can be defined from protein structure alone according to N-glycan site topology. We develop a model describing 2G12 glycoepitopes based on N-glycan site topology, and apply the model to identify viruses within the Protein Data Bank presenting putative 2G12 glycoepitopes for 2G12 repurposing toward analytical, diagnostic, and therapeutic applications.

Mapping Potential Antigenic Drift Sites (PADS) on SARS-CoV-2 Spike in Continuous Epitope-Paratope Space.

SARS-CoV-2 mutations with antigenic effects pose a risk to immunity developed through vaccination and natural infection. While vaccine updates for current variants of concern (VOCs) are underway, it is likewise important to prepare for further antigenic mutations as the virus navigates the heterogeneous global landscape of host immunity. Toward this end, a wealth of data and tools exist that can augment existing genetic surveillance of VOC evolution. In this study, we integrate published datasets describing genetic, structural, and functional constraints on mutation along with computational analyses of antibody-spike co-crystal structures to identify a set of potential antigenic drift sites (PADS) within the receptor binding domain (RBD) and N-terminal domain (NTD) of SARS-CoV-2 spike protein. Further, we project the PADS set into a continuous epitope-paratope space to facilitate interpretation of the degree to which newly observed mutations might be antigenically synergistic with existing VOC mutations, and this representation suggests that functionally convergent and synergistic antigenic mutations are accruing across VOC NTDs. The PADS set and synergy visualization serve as a reference as new mutations are detected on VOCs, enable proactive investigation of potentially synergistic mutations, and offer guidance to antibody and vaccine design efforts.

Glycans in Virus-Host Interactions: A Structural Perspective.

Many interactions between microbes and their hosts are driven or influenced by glycans, whose heterogeneous and difficult to characterize structures have led to an underappreciation of their role in these interactions compared to protein-based interactions. Glycans decorate microbe glycoproteins to enhance attachment and fusion to host cells, provide stability, and evade the host immune system. Yet, the host immune system may also target these glycans as glycoepitopes. In this review, we provide a structural perspective on the role of glycans in host-microbe interactions, focusing primarily on viral glycoproteins and their interactions with host adaptive immunity. In particular, we discuss a class of topological glycoepitopes and their interactions with topological mAbs, using the anti-HIV mAb 2G12 as the archetypical example. We further offer our view that structure-based glycan targeting strategies are ready for application to viruses beyond HIV, and present our perspective on future development in this area.

Insights on the mutational landscape of the SARS-CoV-2 Omicron variant.

The SARS-COV2 Omicron variant has sparked global concern due to the possibility of enhanced transmissibility and escape from vaccines and therapeutics. In this study, we describe the mutational landscape of the Omicron variant using amino acid interaction (AAI) networks. AAI network analysis is particularly well suited for interrogating the impact of constellations of mutations as occur on Omicron that may function in an epistatic manner. Our analyses suggest that as compared to previous variants of concern, the Omicron variant has increased antibody escape breadth due to mutations in class 3 and 4 antibody epitopes as well as increased escape depth due to accumulated mutations in class 1 antibody epitopes. We note certain RBD mutations that might further enhance Omicron's escape, and in particular advise careful surveillance of two subclades bearing R346S/K mutations with relevance for certain therapeutic antibodies. Further, AAI network analysis suggests that the function of certain therapeutic monoclonal antibodies may be disrupted by Omicron mutations as a result of the cumulative indirect perturbations to the epitope surface properties, despite point-mutation analyses suggesting these antibodies are tolerant of the set of Omicron mutations in isolation. Finally, for several Omicron mutations that do not appear to contribute meaningfully to antibody escape, we find evidence for a plausible role in enhanced transmissibility via disruption of RBD-down conformational stability at the RBD-RBD interface.

An Antigenic Space Framework for Understanding Antibody Escape of SARS-CoV-2 Variants.

The evolution of mutations in SARS-CoV-2 at antigenic sites that impact neutralizing antibody responses in humans poses a risk to immunity developed through vaccination and natural infection. The highly successful RNA-based vaccines have enabled rapid vaccine updates that incorporate mutations from current variants of concern (VOCs). It is therefore important to anticipate future antigenic mutations as the virus navigates the heterogeneous global landscape of host immunity. Toward this goal, we survey epitope-paratope interfaces of anti-SARS-CoV-2 antibodies to map an antigenic space that captures the role of each spike protein residue within the polyclonal antibody response directed against the ACE2-receptor binding domain (RBD) or the N-terminal domain (NTD). In particular, the antigenic space map builds on recently published epitope definitions by annotating epitope overlap and orthogonality at the residue level. We employ the antigenic space map as a framework to understand how mutations on nine major variants contribute to each variant's evasion of neutralizing antibodies. Further, we identify constellations of mutations that span the orthogonal epitope regions of the RBD and NTD on the variants with the greatest antibody escape. Finally, we apply the antigenic space map to predict which regions of antigenic space-should they mutate-may be most likely to complementarily augment antibody evasion for the most evasive and transmissible VOCs.

Raman Spectroscopy for Rapid Evaluation of Surgical Margins during Breast Cancer Lumpectomy.

Failure to precisely distinguish malignant from healthy tissue has severe implications for breast cancer surgical outcomes. Clinical prognoses depend on precisely distinguishing healthy from malignant tissue during surgery. Laser Raman spectroscopy (LRS) has been previously shown to differentiate benign from malignant tissue in real time. However, the cost, assembly effort, and technical expertise needed for construction and implementation of the technique have prohibited widespread adoption. Recently, Raman spectrometers have been developed for non-medical uses and have become commercially available and affordable. Here we demonstrate that this current generation of Raman spectrometers can readily identify cancer in breast surgical specimens. We evaluated two commercially available, portable, near-infrared Raman systems operating at excitation wavelengths of either 785 nm or 1064 nm, collecting a total of 164 Raman spectra from cancerous, benign, and transitional regions of resected breast tissue from six patients undergoing mastectomy. The spectra were classified using standard multivariate statistical techniques. We identified a minimal set of spectral bands sufficient to reliably distinguish between healthy and malignant tissue using either the 1064 nm or 785 nm system. Our results indicate that current generation Raman spectrometers can be used as a rapid diagnostic technique distinguishing benign from malignant tissue during surgery.